LetTokenBurnBenchHub

Model

Claude Opus 5

Anthropic · closed weights

Knowledge & Reasoning

Knowledge & Reasoning

Benchmark	Score	Metric	Harness	Source	Confidence	Link
ArcticAbstain	44.4	abstention_rate	N/A	Independent	medium	arxiv.org
Balrog	63.4	average_progress	N/A	Independent	medium	balrogai.com
CritPt	29.1	accuracy	N/A	Independent	low	epoch.ai
FREA	68.0	balanced_accuracy	N/A	Independent	medium	arxiv.org
GPQA Diamond	93.9	accuracy	N/A	Independent	high	epoch.ai
HouseholdBench	0.9	relmae	N/A	Independent	medium	arxiv.org
MOF-VERIFY	62.8	macro-f1	N/A	Independent	medium	arxiv.org
SDDBench	92.9	accuracy	N/A	Independent	medium	arxiv.org
SimpleQA Verified	59.9	accuracy	N/A	Independent	high	epoch.ai
TasteVal	2.1	compute_multiplier	N/A	Independent	medium	arxiv.org

Coding

Benchmark	Score	Metric	Harness	Source	Confidence	Link
A2Z GameSpec-Bench	73.9	gdd_fidelity	N/A	Independent	medium	arxiv.org
ArchitectureIQ	76.0	accuracy	architectureiq	Independent	medium	arxiv.org
ArrivalBench	16.7	failure_rate	N/A	Independent	medium	arxiv.org
CRBugBench	48.3	success_rate	N/A	Independent	medium	arxiv.org
DeepSWE	73.6	pass@1	mini-swe-agent	Independent	medium	deepswe.datacurve.ai
E2E-SWE	67.7	pass@1	N/A	Independent	medium	arxiv.org
ModeBench	0.1	pcmd	N/A	Independent	medium	arxiv.org
ParanoiaEval	58.7	violation_rate	N/A	Independent	medium	arxiv.org
SciCode	56.4	score	N/A	Independent	medium	artificialanalysis.ai
TestSuiteBench	55.5	score	N/A	Independent	medium	arxiv.org
WorldSolver	46.7	score	N/A	Independent	medium	arxiv.org

Agentic / Tool Use

Agentic / Tool Use

Benchmark	Score	Metric	Harness	Source	Confidence	Link
BOTTLED	0.4	f1	N/A	Independent	medium	arxiv.org
BrickBench	1101.0	elo	N/A	Independent	medium	arxiv.org
cua-speedrun	91.6	success_rate	N/A	Independent	medium	arxiv.org
DataSense-Bench	35.9	accuracy	N/A	Independent	medium	arxiv.org
DAYJOB	8.4	success_rate	N/A	Independent	medium	arxiv.org
DrugTargetWorld	14.8	score	N/A	Independent	medium	arxiv.org
EngiWorld	44.3	engiscore	N/A	Independent	medium	arxiv.org
Incident-Arena	0.4	pass@1	claude code	Independent	medium	arxiv.org
KlinikeBench	29.4	strict_pass@1	N/A	Independent	medium	arxiv.org
Learn2Play Bench	73.0	max	claude code	Independent	medium	arxiv.org
Learn2Play Bench	54.7	max	evotest	Independent	medium	arxiv.org
Learn2Play Bench	50.3	max	opencode	Independent	medium	arxiv.org
Learn2Play Bench	46.2	max	memory	Independent	medium	arxiv.org
Mine Odyssey	50.6	success_rate	N/A	Independent	medium	arxiv.org
MMPostTrainBench	44.4	mean_task_score_(%)	claude code	Independent	medium	arxiv.org
Open-Endedness Bench	0.8	e4_no_reward_hacking	N/A	Independent	medium	arxiv.org
TALK-Dem	96.7	success_rate	N/A	Independent	medium	arxiv.org
WorldAuditBench	28.2	success_rate	vlm auditors	Independent	medium	arxiv.org
WorldAuditBench	15.0	success_rate	vla–vlm auditors	Independent	medium	arxiv.org

Multimodal

Benchmark	Score	Metric	Harness	Source	Confidence	Link
GDP.pdf	24.0	gdp.pdf_score	N/A	Independent	medium	surgehq.ai
Knossos	70.4	f1	N/A	Independent	medium	arxiv.org
PlotGround-1k	60.2	accuracy	N/A	Independent	medium	arxiv.org
Video-Index	56.8	accuracy	N/A	Independent	medium	arxiv.org

Safety & Alignment

Safety & Alignment

Benchmark	Score	Metric	Harness	Source	Confidence	Link
APEX-Harm	6.5	success_rate	N/A	Independent	medium	arxiv.org
DecepEval	36.0	deception_rate	N/A	Independent	medium	arxiv.org
DRYLAB-Bench	0.1	floor-included_brs_s2	N/A	Independent	medium	arxiv.org
EvoRiskBench	43.4	success_rate	N/A	Independent	medium	arxiv.org
Nonpublic Vulnerabilities Benchmark	0.6	attack_score	claude code	Independent	medium	arxiv.org

Abstract Reasoning

Abstract Reasoning

Benchmark	Score	Metric	Harness	Source	Confidence	Link
ARC-AGI-2	90.4	score	N/A	Independent	low	epoch.ai

Reproduced by the community